Project Summary/Abstract Symptoms of insomnia and anxiety are common in patients with chronic obstructive pulmonary disease (COPD) and detract from quality of life. These symptoms are further exacerbated in the setting of mental health disorders such as posttraumatic stress disorder (PTSD). Although patients with COPD and comorbid PTSD may experience temporary relief from the use of benzodiazepines, these medications pose significant risks. In patients with COPD, benzodiazepine use is associated with a 45% increased risk of exacerbations, and increased risks for respiratory failure and mortality?although our understanding of these risks is incomplete. In patients with PTSD, benzodiazepine use reinforces PTSD symptoms, impairs recovery, and increases the risk of subsequent substance abuse. Given these findings, guidelines have advocated against the routine use of benzodiazepines in patients with either COPD or PTSD, although there may be utility for use in end-of-life symptom palliation. Despite these recommendations, we found that benzodiazepines are used in 35% of patients with COPD and comorbid PTSD in the Veteran's Health Administration (VA). This rate of use is approximately seven times that observed in the general population. We hypothesize that system level factors such as lower center volume, and lower complexity of care provided will be associated with benzodiazepine use after adjustment for patient level factors. Furthermore, as each condition predisposes patients to unique risks related to benzodiazepine use, we hypothesize that the use of benzodiazepines in patients with COPD and comorbid PTSD will result in excess mortality. The current project seeks to test these hypotheses in three aims: 1) identify predictors of chronic benzodiazepine use in a national population of patients with COPD and comorbid PTSD, 2) evaluate mortality risk related to chronic benzodiazepine use in patients with COPD and comorbid PTSD in a propensity-matched cohort, and 3) identify patient and provider perspectives regarding the use of benzodiazepine medications to reduce anxiety and insomnia symptoms in patients with COPD and comorbid PTSD. Knowledge of predictors and perspectives regarding benzodiazepine use gained from accomplishing aims 1 and 3 will inform targets for de-implementation of guideline inconsistent benzodiazepine use, and knowledge gained from aim 2 will help gauge the potential mortality impact of such an intervention. Future efforts aimed at de-implementation will focus on the substitution of alternative agents and education regarding the harms of benzodiazepine use in this setting.